Biomechanical modelling of soft tissue provides a method for constraining medical image registration, such that the estimated spatial transformation is considered biophysically plausible. Existing methods either directly optimize the loss function containing the biomechanical-constrained regularization term over deformations, which takes much computational time, or are trained using biomechanically plausible data generated via finite element simulation, which is cumbersome. This work first instantiates the recently-proposed physics-informed neural networks (PINNs) to 3D elastic models that are used to establish the partial differential equations (PDEs) representing physics laws of biomechanical constraints to be satisfied. The registration algorithm that aligns point sets considering PINN-imposed biomechanics (i.e., the forward problem) is then formulated. In addition, the inverse problem and its algorithm of physical parameter (i.e., material property) estimation along with the registration are also formulated and developed. We carefully compare linear and nonlinear elasticity theories' capabilities in solving both tasks of forward registration and inverse physical parameter identification under PINNs respectively. Furthermore, two specific network configurations that leverage one common branch or two individual branches to predict deformation vectors and biomechanical states are also constructed and compared. The proposed PINNs-based registration approaches have been extensively evaluated with three experiments, that is single and multiple patient MRI-US registration using clinical MRI-US pairs, and registration using pairs of undeformed MR images from clinical cases of prostate cancer biopsy and deformed counterparts with finite-element-computed ground-truth deformation. Results demonstrate that the proposed methods achieve state-of-the-art performances compared to biomechanical-model-based and learning-based registration approaches, and the biomechanical constraints of soft tissues have been successfully warranted after registration. The codes are available at https://github.com/ZheMin-1992/Registration_PINNs.

The prowess that makes few-shot learning desirable in medical image analysis is the efficient use of the support image data, which are labelled to classify or segment new classes, a task that otherwise requires substantially more training images and expert annotations. This work describes a fully 3D prototypical few-shot segmentation algorithm, such that the trained networks can be effectively adapted to clinically interesting structures that are absent in training, using only a few labelled images from a different institute. First, to compensate for the widely recognised spatial variability between institutions in episodic adaptation of novel classes, a novel spatial registration mechanism is integrated into prototypical learning, consisting of a segmentation head and an spatial alignment module. Second, to assist the training with observed imperfect alignment, support mask conditioning module is proposed to further utilise the annotation available from the support images. Extensive experiments are presented in an application of segmenting eight anatomical structures important for interventional planning, using a data set of 589 pelvic T2-weighted MR images, acquired at seven institutes. The results demonstrate the efficacy in each of the 3D formulation, the spatial registration, and the support mask conditioning, all of which made positive contributions independently or collectively. Compared with the previously proposed 2D alternatives, the few-shot segmentation performance was improved with statistical significance, regardless whether the support data come from the same or different institutes.

Convolutional neural networks (CNNs) have recently led to significant advances in automatic segmentations of anatomical structures in medical images, and a wide variety of network architectures are now available to the research community. For applications such as segmentation of the prostate in magnetic resonance images (MRI), the results of the PROMISE12 online algorithm evaluation platform have demonstrated differences between the best-performing segmentation algorithms in terms of numerical accuracy using standard metrics such as the Dice score and boundary distance. These small differences in the segmented regions/boundaries outputted by different algorithms may potentially have an unsubstantial impact on the results of downstream image analysis tasks, such as estimating organ volume and multimodal image registration, which inform clinical decisions. This impact has not been previously investigated. In this work, we quantified the accuracy of six different CNNs in segmenting the prostate in 3D patient T2-weighted MRI scans and compared the accuracy of organ volume estimation and MRI-ultrasound (US) registration errors using the prostate segmentations produced by different networks. Networks were trained and tested using a set of 232 patient MRIs with labels provided by experienced clinicians. A statistically significant difference was found among the Dice scores and boundary distances produced by these networks in a non-parametric analysis of variance (p < 0.001 and p < 0.001, respectively), where the following multiple comparison tests revealed that the statistically significant difference in segmentation errors were caused by at least one tested network. Gland volume errors (GVEs) and target registration errors (TREs) were then estimated using the CNN-generated segmentations. Interestingly, there was no statistical difference found in either GVEs or TREs among different networks, (p = 0.34 and p = 0.26, respectively). This result provides a real-world example that these networks with different segmentation performances may potentially provide indistinguishably adequate registration accuracies to assist prostate cancer imaging applications. We conclude by recommending that the differences in the accuracy of downstream image analysis tasks that make use of data output by automatic segmentation methods, such as CNNs, within a clinical pipeline should be taken into account when selecting between different network architectures, in addition to reporting the segmentation accuracy.

One of the fundamental challenges in supervised learning for multimodal image registration is the lack of ground-truth for voxel-level spatial correspondence. This work describes a method to infer voxel-level transformation from higher-level correspondence information contained in anatomical labels. We argue that such labels are more reliable and practical to obtain for reference sets of image pairs than voxel-level correspondence. Typical anatomical labels of interest may include solid organs, vessels, ducts, structure boundaries and other subject-specific ad hoc landmarks. The proposed end-to-end convolutional neural network approach aims to predict displacement fields to align multiple labelled corresponding structures for individual image pairs during the training, while only unlabelled image pairs are used as the network input for inference. We highlight the versatility of the proposed strategy, for training, utilising diverse types of anatomical labels, which need not to be identifiable over all training image pairs. At inference, the resulting 3D deformable image registration algorithm runs in real-time and is fully-automated without requiring any anatomical labels or initialisation. Several network architecture variants are compared for registering T2-weighted magnetic resonance images and 3D transrectal ultrasound images from prostate cancer patients. A median target registration error of 3.6 mm on landmark centroids and a median Dice of 0.87 on prostate glands are achieved from cross-validation experiments, in which 108 pairs of multimodal images from 76 patients were tested with high-quality anatomical labels.

Statistical shape models of soft-tissue organ motion provide a useful means of imposing physical constraints on the displacements allowed during non-rigid image registration, and can be especially useful when registering sparse and/or noisy image data. In this paper, we describe a method for generating a subject-specific statistical shape model that captures prostate deformation for a new subject given independent population data on organ shape and deformation obtained from magnetic resonance (MR) images and biomechanical modelling of tissue deformation due to transrectal ultrasound (TRUS) probe pressure. The characteristics of the models generated using this method are compared with corresponding models based on training data generated directly from subject-specific biomechanical simulations using a leave-one-out cross validation. The accuracy of registering MR and TRUS images of the prostate using the new prostate models was then estimated and compared with published results obtained in our earlier research. No statistically significant difference was found between the specificity and generalisation ability of prostate shape models generated using the two approaches. Furthermore, no statistically significant difference was found between the landmark-based target registration errors (TREs) following registration using different models, with a median (95th percentile) TRE of 2.40 (6.19) mm versus 2.42 (7.15) mm using models generated with the new method versus a model built directly from patient-specific biomechanical simulation data, respectively (N = 800; 8 patient datasets; 100 registrations per patient). We conclude that the proposed method provides a computationally efficient and clinically practical alternative to existing complex methods for modelling and predicting subject-specific prostate deformation, such as biomechanical simulations, for new subjects. The method may also prove useful for generating shape models for other organs, for example, where only limited shape training data from dynamic imaging is available.

There is growing clinical demand for image registration techniques that allow multimodal data fusion for accurate targeting of needle biopsy and ablative prostate cancer treatments. However, during procedures where transrectal ultrasound (TRUS) guidance is used, substantial gland deformation can occur due to TRUS probe pressure. In this paper, the ability of a statistical shape/motion model, trained using finite element simulations, to predict and compensate for this source of motion is investigated. Three-dimensional ultrasound images acquired on five patient prostates, before and after TRUS-probe-induced deformation, were registered using a nonrigid, surface-based method, and the accuracy of different deformation models compared. Registration using a statistical motion model was found to outperform alternative elastic deformation methods in terms of accuracy and robustness, and required substantially fewer target surface points to achieve a successful registration. The mean final target registration error (based on anatomical landmarks) using this method was 1.8 mm. We conclude that a statistical model of prostate deformation provides an accurate, rapid and robust means of predicting prostate deformation from sparse surface data, and is therefore well-suited to a number of interventional applications where there is a need for deformation compensation.